Basal Serum Oxytocin Levels in Healthy Non‐Pregnant/Lactating Humans are Below 10 pg/mL and Not Detectable by Traditional Orbitrap LCMS

Abstract

Quantifying peripheral oxytocin (OT) in humans is of increased interest due to its substantial role in cognition and social behavior (e.g. depression, autism). OT is commonly measured by immunoassays, which lack the sensitivity to accurately quantitate the very low basal circulating levels in non‐pregnant/lactating humans due to matrix interferences. This can be resolved by solid phase extraction (SPE) but is inconsistently performed, leading to large variation in OT levels across studies (Leng, 2016). Liquid chromatography mass spectrometry (LCMS) can separate interferences but requires high resolution accurate‐mass (HRAM) capabilities for selective and accurate quantitation.We developed a HRAM orbitrap based LCMS method to measure underivatized native OT from serum after SPE. 0.5 mL serum with OT‐d5 as internal standard underwent C18 SPE followed by Q‐Exactive LCMS analysis with a C18 core‐shell based column (Kinetex, 150 × 3.0 mm; 2.6 μm) and a linear gradient (formic acid/MeCN). The doubly protonated OT ion m/z 562.25503 was selected for MS quantitation due its high signal intensity. Our method LLOQ was 10 pg/mL, which permitted accurate OT measurement in healthy pregnant women (n=4, 16–24 pg/mL), rats (350 pg/mL) and healthy adults after intranasal OT administration. Intra‐day precision was 2–23% and 0% while accuracy was 66~91% and 108~109% at 47.6 and 90.9 pg/mL.A method for ‘total’ OT (OT‐R/A) measuring 500–700 pg/mL in adult plasma was reported recently using LCMSMS after chemical reduction with 1,4‐dithiothreitol, which supposedly disrupted protein binding, followed by alkylation with iodoacetamide to stabilize the reduced product (Brandtzaeg, 2016). Using this method even higher levels were reported from non‐pregnant/lactating subjects (Semba, 2016, Comminos, 2017). However, the masses used for OT‐R/A analysis by Brandtzaeg were for the singly protonated molecule and differed from the theoretical masses by 39.8 ppm while the doubly protonated species used by Semba for quantitation (562.8/839.1(b6); 562.8/286.1(y3); 562.8/446.0(y4) differed by 970 pm. These measured masses were shown in our HRAM LCMS system to be interferences and not OT‐R/A. When we selected a mass window of ≤6 ppm, our HRAM system was unable to detect native OT‐R/A while OT‐R/A was detectable after spiking with OT down to 10 pg/mL plus OT‐d5‐R/A was detected each time after adding OT‐d5 as internal standard.Underivatized OT was correctly measured in monkeys by LCMSMS after SPE purification using MRM transitions 1007‐>723, 285 and 1012‐>723, 290 for the singly protonated native OT and OT‐d5, respectively with basal levels <LLOQ (10 pg/mL; Lee, 2017), which is in excellent agreement with our results. Similarly, others reported mean circulating OT levels <10 pg/mL in non‐pregnant/lactating humans using 2D‐LCMSMS (1–2 pg/mL; Zhang, 2011) with 3 of the 10 subjects <LLOQ (1 pg/mL) and ELISA after SPE (Carson, 2015; mean 6.7 pg/mL) or <2 pg/mL using RIA after SPE (van Londen, 1997; Modahl, 1998). Accurate OT methods find <10 pg/mL OT in serum/plasma of non‐pregnant/lactating/OT medicated humans.PMIDs: Brandtzaeg 27528413; Carson 25349162; Comminos 28112678; Lee 28289281; Leng 27467712; Modahl 9513736; Semba 28508553; van Londen 9326754; Zhang 21609710Support or Funding InformationnoneThis abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.